[
https://issues.apache.org/jira/browse/CALCITE-3621?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Lei Jiang updated CALCITE-3621:
-------------------------------
Description:
JDBC adapter can't push down sort to DB
{code:java}
select ename from scott.emp order by empno
{code}
{code:java}
PLAN=EnumerableSort(sort0=[$1], dir0=[ASC])
JdbcToEnumerableConverter
JdbcProject(ENAME=[$1], EMPNO=[$0])
JdbcTableScan(table=[[SCOTT, EMP]])
{code}
It should be:
{code:java}
PLAN=JdbcToEnumerableConverter
JdbcSort(sort0=[$1], dir0=[ASC])
JdbcProject(ENAME=[$1], EMPNO=[$0])
JdbcTableScan(table=[[SCOTT, EMP]])
{code}
I think the root cause is JdbcSortRule, it convert input's trait to "JDBC,
{color:#ff0000}[1]{color}". that is, input's relset will add a "JDBC, [1]"
subset. But there is nothing rule can convert that input to a rel with "JDBC,
{color:#ff0000}[1]{color}", so EnumerableSort win.
{code:java}
public RelNode convert(Sort sort, boolean convertInputTraits) {
final RelTraitSet traitSet = sort.getTraitSet().replace(out);
final RelNode input;
if (convertInputTraits) {
input = convert(sort.getInput(), traitSet);
} else {
input = sort.getInput();
}
return new JdbcSort(sort.getCluster(), traitSet,
input, sort.getCollation(), sort.offset, sort.fetch);
}
{code}
This is my a part of change: convert input's trait to "JDBC, []"
{code:java}
public RelNode convert(Sort sort, boolean convertInputTraits) {
final RelTraitSet traitSet = sort.getTraitSet().replace(out);
//update again
final RelTraitSet inputTraitSet = sort.getInput().getTraitSet().replace(out);
final RelNode input;
if (convertInputTraits) {
//update
input = convert(sort.getInput(), inputTraitSet);
} else {
input = sort.getInput();
}
return new JdbcSort(sort.getCluster(), traitSet,
input, sort.getCollation(), sort.offset, sort.fetch);
}
{code}
When JdbcSort is used, my solution will trigger a regression issue
JdbcTest#testSelfJoinDifferentColumns(): Project->Sort->Join(left.A, right.A),
Join.left and Join.right have same field A. My solution: select list of Sort
contains explicit column name instead of *
{code:java}
RelToSqlConverter#visit(Sort e)
...
Result x = visitChild(0, e.getInput());
Builder builder = x.builder(e, Clause.ORDER_BY);
//my update
if (builder.select.getSelectList() == null) {
final List<SqlNode> selectList = Expressions.list();
for (RelDataTypeField field : e.getRowType().getFieldList()) {
addSelect(selectList, builder.context.field(field.getIndex()),
e.getRowType());
}
builder.select.setSelectList(new SqlNodeList(selectList, POS));
}
//end
...{code}
Before:
SELECT "t2"."full_name", "t2"."last_name{color:#ff0000}0{color}" AS "last_name"
FROM (SELECT {color:#ff0000}*{color}
FROM (SELECT "last_name"
FROM "foodmart"."employee") AS "t"
INNER JOIN (SELECT "full_name", "first_name", "last_name"
FROM "foodmart"."employee") AS "t0" ON "t"."last_name" = "t0"."first_name"
ORDER BY "t0"."last_name" NULLS LAST
After:
SELECT "t2"."full_name", "t2"."last_name0" AS "last_name"
FROM (SELECT {color:#ff0000}"t"."last_name", "t0"."full_name",
"t0"."first_name", "t0"."last_name" AS "last_name0"{color}
FROM (SELECT "last_name"
FROM "foodmart"."employee") AS "t"
INNER JOIN (SELECT "full_name", "first_name", "last_name"
FROM "foodmart"."employee") AS "t0" ON "t"."last_name" = "t0"."first_name"
ORDER BY "t0"."last_name" NULLS LAST
LIMIT 3) AS "t2"
was:
JDBC adapter can't push down sort to DB
{code:java}
select ename from scott.emp order by empno
{code}
{code:java}
PLAN=EnumerableSort(sort0=[$1], dir0=[ASC])
JdbcToEnumerableConverter
JdbcProject(ENAME=[$1], EMPNO=[$0])
JdbcTableScan(table=[[SCOTT, EMP]])
{code}
It should be:
{code:java}
PLAN=JdbcToEnumerableConverter
JdbcSort(sort0=[$1], dir0=[ASC])
JdbcProject(ENAME=[$1], EMPNO=[$0])
JdbcTableScan(table=[[SCOTT, EMP]])
{code}
I think the root cause is JdbcSortRule, it convert input's trait to "JDBC,
{color:#ff0000}[1]{color}". that is, input's relset will add a "JDBC, [1]"
subset. But there is nothing rule can convert that input to a rel with "JDBC,
{color:#ff0000}[1]{color}", so EnumerableSort win.
{code:java}
public RelNode convert(Sort sort, boolean convertInputTraits) {
final RelTraitSet traitSet = sort.getTraitSet().replace(out);
final RelNode input;
if (convertInputTraits) {
input = convert(sort.getInput(), traitSet);
} else {
input = sort.getInput();
}
return new JdbcSort(sort.getCluster(), traitSet,
input, sort.getCollation(), sort.offset, sort.fetch);
}
{code}
This is my a part of change: convert input's trait to "JDBC, []"
{code:java}
public RelNode convert(Sort sort, boolean convertInputTraits) {
final RelTraitSet traitSet = sort.getTraitSet().replace(out);
//update
RelTraitSet inputTraitSet = traitSet.replace(RelCollations.EMPTY);
final RelNode input;
if (convertInputTraits) {
//update
input = convert(sort.getInput(), inputTraitSet);
} else {
input = sort.getInput();
}
return new JdbcSort(sort.getCluster(), traitSet,
input, sort.getCollation(), sort.offset, sort.fetch);
}
{code}
I have updated some classes to reslove this issue
When JdbcSort is used, my solution will trigger a regression issue
JdbcTest#testSelfJoinDifferentColumns(): Project->Sort->Join(left.A, right.A),
Join.left and Join.right have same field A. My solution: select list of Sort
contains explicit column name instead of *
{code:java}
RelToSqlConverter#visit(Sort e)
...
Result x = visitChild(0, e.getInput());
Builder builder = x.builder(e, Clause.ORDER_BY);
//my update
if (builder.select.getSelectList() == null) {
final List<SqlNode> selectList = Expressions.list();
for (RelDataTypeField field : e.getRowType().getFieldList()) {
addSelect(selectList, builder.context.field(field.getIndex()),
e.getRowType());
}
builder.select.setSelectList(new SqlNodeList(selectList, POS));
}
//end
...{code}
Before:
SELECT "t2"."full_name", "t2"."last_name{color:#ff0000}0{color}" AS "last_name"
FROM (SELECT {color:#ff0000}*{color}
FROM (SELECT "last_name"
FROM "foodmart"."employee") AS "t"
INNER JOIN (SELECT "full_name", "first_name", "last_name"
FROM "foodmart"."employee") AS "t0" ON "t"."last_name" = "t0"."first_name"
ORDER BY "t0"."last_name" NULLS LAST
After:
SELECT "t2"."full_name", "t2"."last_name0" AS "last_name"
FROM (SELECT {color:#ff0000}"t"."last_name", "t0"."full_name",
"t0"."first_name", "t0"."last_name" AS "last_name0"{color}
FROM (SELECT "last_name"
FROM "foodmart"."employee") AS "t"
INNER JOIN (SELECT "full_name", "first_name", "last_name"
FROM "foodmart"."employee") AS "t0" ON "t"."last_name" = "t0"."first_name"
ORDER BY "t0"."last_name" NULLS LAST
LIMIT 3) AS "t2"
> JDBC adapter can't push down sort to DB
> ---------------------------------------
>
> Key: CALCITE-3621
> URL: https://issues.apache.org/jira/browse/CALCITE-3621
> Project: Calcite
> Issue Type: Bug
> Components: jdbc-adapter
> Affects Versions: 1.21.0
> Reporter: Lei Jiang
> Assignee: Lei Jiang
> Priority: Major
> Labels: pull-request-available
> Fix For: 1.22.0
>
> Time Spent: 1.5h
> Remaining Estimate: 0h
>
> JDBC adapter can't push down sort to DB
> {code:java}
> select ename from scott.emp order by empno
> {code}
> {code:java}
> PLAN=EnumerableSort(sort0=[$1], dir0=[ASC])
> JdbcToEnumerableConverter
> JdbcProject(ENAME=[$1], EMPNO=[$0])
> JdbcTableScan(table=[[SCOTT, EMP]])
> {code}
> It should be:
> {code:java}
> PLAN=JdbcToEnumerableConverter
> JdbcSort(sort0=[$1], dir0=[ASC])
> JdbcProject(ENAME=[$1], EMPNO=[$0])
> JdbcTableScan(table=[[SCOTT, EMP]])
> {code}
> I think the root cause is JdbcSortRule, it convert input's trait to "JDBC,
> {color:#ff0000}[1]{color}". that is, input's relset will add a "JDBC, [1]"
> subset. But there is nothing rule can convert that input to a rel with "JDBC,
> {color:#ff0000}[1]{color}", so EnumerableSort win.
> {code:java}
> public RelNode convert(Sort sort, boolean convertInputTraits) {
> final RelTraitSet traitSet = sort.getTraitSet().replace(out);
> final RelNode input;
> if (convertInputTraits) {
> input = convert(sort.getInput(), traitSet);
> } else {
> input = sort.getInput();
> }
> return new JdbcSort(sort.getCluster(), traitSet,
> input, sort.getCollation(), sort.offset, sort.fetch);
> }
> {code}
> This is my a part of change: convert input's trait to "JDBC, []"
> {code:java}
> public RelNode convert(Sort sort, boolean convertInputTraits) {
> final RelTraitSet traitSet = sort.getTraitSet().replace(out);
> //update again
> final RelTraitSet inputTraitSet =
> sort.getInput().getTraitSet().replace(out);
> final RelNode input;
> if (convertInputTraits) {
> //update
> input = convert(sort.getInput(), inputTraitSet);
> } else {
> input = sort.getInput();
> }
> return new JdbcSort(sort.getCluster(), traitSet,
> input, sort.getCollation(), sort.offset, sort.fetch);
> }
> {code}
>
> When JdbcSort is used, my solution will trigger a regression issue
> JdbcTest#testSelfJoinDifferentColumns(): Project->Sort->Join(left.A,
> right.A), Join.left and Join.right have same field A. My solution: select
> list of Sort contains explicit column name instead of *
>
> {code:java}
> RelToSqlConverter#visit(Sort e)
> ...
> Result x = visitChild(0, e.getInput());
> Builder builder = x.builder(e, Clause.ORDER_BY);
> //my update
> if (builder.select.getSelectList() == null) {
> final List<SqlNode> selectList = Expressions.list();
> for (RelDataTypeField field : e.getRowType().getFieldList()) {
> addSelect(selectList, builder.context.field(field.getIndex()),
> e.getRowType());
> }
> builder.select.setSelectList(new SqlNodeList(selectList, POS));
> }
> //end
> ...{code}
> Before:
> SELECT "t2"."full_name", "t2"."last_name{color:#ff0000}0{color}" AS
> "last_name"
> FROM (SELECT {color:#ff0000}*{color}
> FROM (SELECT "last_name"
> FROM "foodmart"."employee") AS "t"
> INNER JOIN (SELECT "full_name", "first_name", "last_name"
> FROM "foodmart"."employee") AS "t0" ON "t"."last_name" = "t0"."first_name"
> ORDER BY "t0"."last_name" NULLS LAST
> After:
> SELECT "t2"."full_name", "t2"."last_name0" AS "last_name"
> FROM (SELECT {color:#ff0000}"t"."last_name", "t0"."full_name",
> "t0"."first_name", "t0"."last_name" AS "last_name0"{color}
> FROM (SELECT "last_name"
> FROM "foodmart"."employee") AS "t"
> INNER JOIN (SELECT "full_name", "first_name", "last_name"
> FROM "foodmart"."employee") AS "t0" ON "t"."last_name" = "t0"."first_name"
> ORDER BY "t0"."last_name" NULLS LAST
> LIMIT 3) AS "t2"
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
